Final Project

Assembly of the SARS-CoV-2 Spike Protein using the de Bruijn Graph
Jonathan Dekraker¹, José Solórzano², Samuel Kitzerow¹

1. Department of Computer Science.
2. Department of Plant Pathology.

Run Main Program

Use default file (sars_spike_protein_reads.fastq):

python .\main.py

User entered parameters:

python .\main.py -l			# Input list indices [start:end]
python .\main.py -k			# Input kmer range
python .\main.py -kl			# Input both

Run Alignment

python .\alignment.py

Run Trimming

java -jar trimmomatic-0.39.jar SE -phred33 ../input/sars_spike_protein_raw_reads.fastq ../output.fastq ILLUMINACLIP:TruSeq3-SE:2:30:10 LEADING:3 TRAILING:3 SLIDINGWINDOW:4:15 MINLEN:36

Running Cleanup

Allows trimming, discards bad reads, and alpha value is 0.1.

./lighter -r ../output.fastq -k 17 5000000 0.1 -t 10 -trim -discard

edges.txt: - File that contains all directed edges. Created by printing set from main.py

eulerianCycle.py: - Computes and prints eulerian circuit from spike_protein_directed_graph.txt

eulerianPath.py: - Computes and prints eulerian path from spike_protein_directed_graph.txt

main.py: - Only change is the create_directed_graph function which creates spike_protein_directed_graph.txt

spike_protein_directed_graph.txt: - A directed graph where the graph is given in the form of an adjacency list (from edges.txt)

Function Runtimes

The system becomes sluggish when k-mers greater than 5 are used. The addition of progress bars helps judge the completion of the program; however, the image function that creates the directed graph is slow. Upon investigation it was determined that nx.draw(fig, pos, **options) and if(save_fig): plt.savefig(file, dpi=500) takes a considerable amount of time (with the later taking the longest to complete).